1. Field of the Invention
The present invention relates to an image processing apparatus and, more particularly, to an image processing apparatus which makes it possible to execute the processes such as movement, enlargement/reduction with a variable magnification, rotation, and the like of digital image information.
2. Related Background Art
Hitherto, the affine transformation has been known as a method for accomplishing the rotation and the like of an image; Expression (1) is a general expression of the two-dimensional affine transformation. Assuming that the coordinates of an arbitrary point of the original image are (S.sub.x, S.sub.y) and the coordinates after the transformation of the corresponding point are (D.sub.x, D.sub.y), expression (1) can be expressed as follows. ##EQU1## The second term on the right side of expression (1) is the transformation matrix. In the case of rotating by only an angle .theta. around an arbitrary point (P.sub.x, P.sub.y), the transformation matrix can be described as follows. ##EQU2##
An image can be also moved and enlarged or reduced with a variable magnification using expression (1).
For example, when the original image is increased n times and moved by the distance t.sub.x in the direction of the x axis and by the distance t.sub.y in the direction of the y axis, the matrix of the second term of the right side of expression (1) will become ##EQU3##
The above method is well known and can be realized by some software or hardware. However, to process the affine transformation at a high speed, it is necessary to realize the above method by hardware. At the same time, the memory to store the image before and after the conversion must be the real memory. This is because it takes a time to input and output the image data before and after the conversion, so that the capability of the hardware to realize expression (1) cannot be effectively used. Therefore, in the conventional image processing apparatus in which the affine transformation is realized by hardware, the processing region is up to about 512.times.512 dots which can be realized by a real memory. Further, random access memories of a full page need to be arranged at the stages before and after the conversion, so that two memories each consisting of 512.times.512 dots are necessary. Thus, there is the drawback such that the cost increases. Moreover, as will be obvious from expression (1), there is also the drawback such that six high speed multipliers are needed to obtain (D.sub.x, D.sub.y).